OpenAI Telegram Bot

This is a repository for a chatbot built using OpenAI's chat completion API. It allows users to ask the bot questions and receive human-like responses from the AI model.

Prerequisites

1. Get OpenAI API key.

2. Get Telegram bot token.

3. Install Docker.

   Here is command for installation on Debian/Ubuntu via convenience script:

   sudo apt update && \
   curl -fsSL https://get.docker.com -o get-docker.sh && \
   /bin/bash ./get-docker.sh && \
   sudo apt install -y docker-compose

How to run locally

1. Clone this repo onto your local machine:

   git clone https://github.com/Danand/bot-openai.git

2. Create .env file with the necessary variables:

   LOG_LEVEL="DEBUG"
   
   TELEGRAM_API_TOKEN="<your-token-here>"
   TELEGRAM_WHITELISTED_USERS="<telegram-user-id-integer-of-admins-comma-separated>"
   
   OPENAI_API_KEY="<your-token-here>"
   OPENAI_DEFAULT_MODEL="gpt-3.5-turbo"
   OPENAI_DEFAULT_TEMPERATURE="1.0"
   OPENAI_DEFAULT_MAX_TOKENS="4097"
   OPENAI_DEFAULT_MAX_MESSAGES="6"
   
   REDIS_HOST="bot-openai-redis"
   REDIS_PORT=6380
   REDIS_VERSION="7.0.11-alpine3.18"

3. Run the bot using the command:

   docker-compose \
     -f ./docker-compose.yml \
     -f ./docker-compose-local.yml \
     up

How to deploy

1. Fork this repo to your own account.

2. Set up necessary secrets in your secrets settings of your repo:

   • DEPLOY_SSH_KEY
   • HOST_PUBLIC_IP
   • OPENAI_API_KEY
   • TELEGRAM_API_TOKEN
   • TELEGRAM_WHITELISTED_USERS
   • _GITHUB_PAT

3. Make changes if you want to.

4. Make sure you have compatible GitHub Actions Runner for this repo.

5. Push new tag, workflow Build and Push Docker Images will be triggered automatically.

6. Run workflow Deploy Docker Images manually with chosen tag to deploy image built at previous step.